Apache
/
commons-math
mirror of https://github.com/apache/commons-math.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

[MATH-1221] Improve performance of ZipfDistribution by caching the nth generalized harmonic.

This commit is contained in:
Thomas Neidhart 2015-05-01 14:12:44 +02:00
parent 002276ea31
commit bd5afc0b5a
2 changed files with 12 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
src/changes/changes.xml
View File

@ -54,6 +54,9 @@ If the output is not quite correct, check for invisible trailing spaces!
</release> </release>
<release version="4.0" date="XXXX-XX-XX" description=""> <release version="4.0" date="XXXX-XX-XX" description="">
<action dev="tn" type="fix" issue="MATH-1221">
Improve performance of "ZipfDistribution" by caching the nth generalized harmonic.
</action>
<action dev="tn" type="fix" issue="MATH-1220" due-to="Otmar Ertl"> <!-- backported to 3.6 --> <action dev="tn" type="fix" issue="MATH-1220" due-to="Otmar Ertl"> <!-- backported to 3.6 -->
Improve performance of "ZipfDistribution#sample()" by using a rejection algorithm. Improve performance of "ZipfDistribution#sample()" by using a rejection algorithm.
</action> </action>

15
src/main/java/org/apache/commons/math4/distribution/ZipfDistribution.java
View File

@ -30,11 +30,13 @@ import org.apache.commons.math4.util.FastMath;
*/ */
public class ZipfDistribution extends AbstractIntegerDistribution { public class ZipfDistribution extends AbstractIntegerDistribution {
/** Serializable version identifier. */ /** Serializable version identifier. */
private static final long serialVersionUID = -140627372283420404L; private static final long serialVersionUID = 20150501L;
/** Number of elements. */ /** Number of elements. */
private final int numberOfElements; private final int numberOfElements;
/** Exponent parameter of the distribution. */ /** Exponent parameter of the distribution. */
private final double exponent; private final double exponent;
/** Cached values of the nth generalized harmonic. */
private final double nthHarmonic;
/** Cached numerical mean */ /** Cached numerical mean */
private double numericalMean = Double.NaN; private double numericalMean = Double.NaN;
/** Whether or not the numerical mean has been calculated */ /** Whether or not the numerical mean has been calculated */
@ -93,6 +95,7 @@ public class ZipfDistribution extends AbstractIntegerDistribution {
this.numberOfElements = numberOfElements; this.numberOfElements = numberOfElements;
this.exponent = exponent; this.exponent = exponent;
this.nthHarmonic = generalizedHarmonic(numberOfElements, exponent);
} }
/** /**
@ -120,7 +123,7 @@ public class ZipfDistribution extends AbstractIntegerDistribution {
return 0.0; return 0.0;
} }
return (1.0 / FastMath.pow(x, exponent)) / generalizedHarmonic(numberOfElements, exponent); return (1.0 / FastMath.pow(x, exponent)) / nthHarmonic;
} }
/** {@inheritDoc} */ /** {@inheritDoc} */
@ -130,7 +133,7 @@ public class ZipfDistribution extends AbstractIntegerDistribution {
return Double.NEGATIVE_INFINITY; return Double.NEGATIVE_INFINITY;
} }
return -FastMath.log(x) * exponent - FastMath.log(generalizedHarmonic(numberOfElements, exponent)); return -FastMath.log(x) * exponent - FastMath.log(nthHarmonic);
} }
/** {@inheritDoc} */ /** {@inheritDoc} */
@ -142,7 +145,7 @@ public class ZipfDistribution extends AbstractIntegerDistribution {
return 1.0; return 1.0;
} }
return generalizedHarmonic(x, exponent) / generalizedHarmonic(numberOfElements, exponent); return generalizedHarmonic(x, exponent) / nthHarmonic;
} }
/** /**
@ -174,7 +177,7 @@ public class ZipfDistribution extends AbstractIntegerDistribution {
final double s = getExponent(); final double s = getExponent();
final double Hs1 = generalizedHarmonic(N, s - 1); final double Hs1 = generalizedHarmonic(N, s - 1);
final double Hs = generalizedHarmonic(N, s); final double Hs = nthHarmonic;
return Hs1 / Hs; return Hs1 / Hs;
} }
@ -210,7 +213,7 @@ public class ZipfDistribution extends AbstractIntegerDistribution {
final double Hs2 = generalizedHarmonic(N, s - 2); final double Hs2 = generalizedHarmonic(N, s - 2);
final double Hs1 = generalizedHarmonic(N, s - 1); final double Hs1 = generalizedHarmonic(N, s - 1);
final double Hs = generalizedHarmonic(N, s); final double Hs = nthHarmonic;
return (Hs2 / Hs) - ((Hs1 * Hs1) / (Hs * Hs)); return (Hs2 / Hs) - ((Hs1 * Hs1) / (Hs * Hs));
} }